Snapped spindle

[roubaixtuesday]

Chris Hoy at his peak would produce 2,500 watts

Yet it was little known, middle aged me whose torque broke Hollowtech.

IN YOUR FACE, HOY BOY!!!

 

[Paulus]

I did once use a hammer and chisel to undo a corroded locknut on an adjustable bearing cup.
None of this hollowtech technical stuff in my shed though.😁

 

[classic33]

Following Silva's earlier example I filled in the as documented actual figure for the HollowTech 2 spindles diameter:

https://amesweb.info/Torsion/torsion-of-shaft-calculator.aspx
Selected "Hollow shaft"
Filled again in some rough bicycle alike values.
HollowTech axle diameter is documented as 24 mm.

Torque [T] 500
Rotation speed [ω] 60 rpm
Shaft outer radius [c2] 12 mm
Shaft inner radius [c1] 11 mm
Shaft length [L] 100
Modulus of rigidity [G] 300 GPa
= 1 mm wall thickness
RESULTS:
Maximum shear stress [τmax] 626.697 MPa
Angle of twist [θ] 0.997°

Torque [T] 500
Rotation speed [ω] 60 rpm
Shaft outer radius [c2] 12 mm
Shaft inner radius [c1] 10.5 mm
Shaft length [L] 100
Modulus of rigidity [G] 300 GPa
= 1.5 mm wall thickness
RESULTS:
Maximum shear stress [τmax] 445.14 MPa
Angle of twist [θ] 0.708°

Maximum shear stress difference is 626.697-445.14 = 181.557
That's 29% less stress in the material.
29% more reserve to avoid a snap, by a 1.5 mm instead of 1 mm thick wall.
The weight "penalty" is:
https://www.profilebendingmachine.com/steel-pipe-tube-hss-weight-calculator/
24 mm outer diameter, wall thickness 1 mm
Result: 60 gram
24 mm outer diameter, wall thickness 1.5 mm
Result: 80 gram

So, for the HollowTech 2 spindle (outer)diameter specification of 24 mm,
Increasing a wall thickness case of 1 mm to 1.5 mm (=+50% thickness)
- gains 29% more resistance to torsion-snapping (=Good For Safety)
- gains 33% more weight (=Bad For Cheating in Races - the "extra" is 20 gram.
Yes, this Spindle Snap Safety +29% increase is Priced as 20 gram, but at the Counter they Complain to the Cashier because the lesser cheating might cost them 3 sec on 3 km and come in second...

... just to illustrate the rather big difference in priority between a cheat-racer and a commuter/shopper that happily wants to trade that 3 sec for no spindle snap.

I get it's not for you, but you realise a 1/3 of the posts on this thread are nowt but you complaining, or trying to prove why they're not what you'd use any more. Just find something that works for you.

Fine, don't like them don't use them. But the flogging of the dead horse has to stop sometime.

 

[Yellow Saddle]

Yet it was little known, middle aged me whose torque broke Hollowtech.

IN YOUR FACE, HOY BOY!!!

Yes but how do we know you weren't all juiced up?

 

[Ming the Merciless]

Yes but how do we know you weren't all juiced up?

@roubaixtuesday Is known as the hammer and chisel of cycling.

 

[roubaixtuesday]

@roubaixtuesday Is known as the hammer and chisel of cycling.

I ride for the Camborne school of mines team

 

[Ajax Bay]

above was Octalink 1, let's go to Octalink 2,

Just a quack one @silva - could I confirm you've never actually fitted a hollowtech2 BB/chainset?
(The OP 'snapped' the spindle of one of this type.)

 

[roubaixtuesday]

Just a quack one

Apposite

 

[Dogtrousers]

Apposite

Eider thought so too.

 

[roubaixtuesday]

Eider thought so too.

Not another thread descending into ruddy puns surely!

 

[Ming the Merciless]

Eider thought so too.

Did you check that on Duck Duck Gone?

 

[Dogtrousers]

Not another thread descending into ruddy puns surely!

Better than the fowl snipes that made up most of the thread.

 

[silva]

Chris Hoy at his peak would produce 2,500 watts at 130-135 rpm sprinting round the velodrome. That only comes to 184 Nm of torque. At the more modest outputs of us mortals it will be far less. A typical cyclist peaks no more than 40 Nm

The power is irrelevant in the comparison since it's the same on both sides of it.
The spindle wall thickness, not, it's 1 mm versus 1.5 mm.
Being the single difference, that causes the material stress difference mentioned and repeated here:
Increasing a wall thickness case of 1 mm to 1.5 mm (=+50% thickness)
- gains 29% more resistance to torsion-snapping (=Good For Safety)
- gains 33% more weight (=Bad For Cheating in Races - the "extra" is 20 gram.

That's obvious, yet you react on the irrelevant power parameter, and don'r react on the relevant 29% extra strength to resist break, at a cost of 33% higher spindle weight. Those figures are Far Far Faaaar Awaaaay from the here worded "minimal" extra strength and "maximal" extra weight, to claim a wall thickness increase as useless. The figures prove that it does make a substantial difference, at an acceptable weight penalty. 20 gram extra traded for 29% less prone to snapping.

 

[roubaixtuesday]

The power is irrelevant in the comparison since it's the same on both sides of it.
The spindle wall thickness, not, it's 1 mm versus 1.5 mm.
Being the single difference, that causes the material stress difference mentioned and repeated here:
Increasing a wall thickness case of 1 mm to 1.5 mm (=+50% thickness)
- gains 29% more resistance to torsion-snapping (=Good For Safety)
- gains 33% more weight (=Bad For Cheating in Races - the "extra" is 20 gram.

That's obvious, yet you react on the irrelevant power parameter, and don'r react on the relevant 29% extra strength to resist break, at a cost of 33% higher spindle weight. Those figures are Far Far Faaaar Awaaaay from the here worded "minimal" extra strength and "maximal" extra weight, to claim a wall thickness increase as useless. The figures prove that it does make a substantial difference, at an acceptable weight penalty. 20 gram extra traded for 29% less prone to snapping.

A bit like the difference between a goosander and a shellduck then?

 

[Ming the Merciless]

The power is irrelevant in the comparison since it's the same on both sides of it.

You put a torque value into the calculator that is roughly 3 times the amount a gold medal winning Olympic track cyclist will produce for seconds. Of course its relevant, put in a torque value it will be subjected to. Try a value of 40 not 500 and see what it comes up with.

hmm, angle of twist 0, max shear stress 1.16